the fomo leaderboard's four clocks
switching the timeframe does not filter the same list. it answers a different question, and three of the four questions are ones you probably did not mean to ask.
fomo's leaderboard offers 24 hour, 7 day, 30 day and all-time views. it looks like one ranking with a filter on it. it is closer to four different rankings that happen to share a page, and each one is dominated by a different effect.
24 hours: mostly noise
on memecoin-sized volatility, a single position can produce more pnl in an afternoon than a disciplined trader generates in a month. the 24 hour board is therefore, to a first approximation, a ranking of who held today's biggest winner.
that is not a repeatable skill, and by the time it is displayed, the move that put them there has already happened. the practical failure is specific: people find a trader at the top of the 24h board, follow them, and discover they have bought exposure to whatever that person does next — which has no particular relationship to the thing that got them ranked.
the 24h board is interesting as a discovery surface for tokens. it is nearly useless as a selection tool for traders.
7 day and 30 day: where the signal lives
these are the two worth reading, and specifically worth reading against each other.
| 7d rank vs 30d rank | what it usually means |
|---|---|
| high on both | consistent recent form — the closest thing to a real signal here |
| high 7d, absent 30d | a hot week, or one large trade. not yet evidence |
| absent 7d, high 30d | cooled off, or stopped trading. find out which before copying |
| middling on both | often the most honest record on the page |
the middling-on-both case deserves more attention than it gets. a trader who is consistently in the upper-middle across both windows is showing you something a spike never does: a process that survives more than one market condition.
all-time: a monument, not a forecast
the all-time board is the one everyone sorts by and the one that answers the least useful question.
two problems compound. the first is that it is dominated by whoever caught the largest move in the platform's history, which may have been a year ago and may never repeat. the second, and worse, is survivorship: it ranks realised pnl among accounts that are still there. it structurally cannot show you the accounts that ran the identical strategy, lost, and stopped.
if a thousand people run a high-variance strategy with no edge whatsoever, the top of the all-time board will still look like genius. the board cannot distinguish that world from one where the strategy works, and neither can you, from the board alone.
the column that is not a timeframe
every timeframe above ranks the past. there is a column on the same page that behaves more like a leading indicator, and almost nobody sorts by it.
we measured 29 first buys from one large fomo wallet and found the move was larger when the trader supplied a smaller share of that minute's volume — the opposite of what their own price impact would produce, and consistent with the followers arriving through the notification being the actual volume.
if that holds, then follower count predicts what a trader's next entry does to a price better than any pnl window does. a trader with half a million followers and average returns moves a token harder than one with forty thousand followers and excellent ones. the sample is small and the caveats are real, but the direction is worth knowing when you read the page.
how to actually read the page
- start on 30d, not all-time and not 24h.
- cross-check against 7d. present on both is the filter that removes most of the noise.
- open the history. count the trades. one big win is not a record.
- look at the follower column separately, as a different question: how much force does their next alert carry?
- check hold time against your own situation. a scalper and a multi-day holder need completely different account sizes to copy, for fee reasons that have nothing to do with skill.
the full selection framework, including the red flags that end the search, is in which fomo traders are worth copying.
frequently asked
which fomo leaderboard timeframe is most useful?
7d and 30d together, read against each other. 24h is dominated by single lucky trades and is close to pure noise for selection purposes. all-time is dominated by whoever caught the biggest move in the platform's history and tells you almost nothing about current form. the 7d-versus-30d comparison is where you can actually see whether someone is consistent or just hot.
why do the top traders change every day on the 24h board?
because a single memecoin position can produce a larger pnl in one afternoon than a careful trader makes in a month. the 24h board is therefore mostly a ranking of who held the day's biggest winner, which is not a repeatable skill and not something you can act on after the fact.
is the all-time leaderboard survivorship-biased?
heavily. it ranks realised pnl among accounts that are still trading and still visible. it cannot show you the accounts that ran an identical strategy, lost, and stopped — and in a high-variance activity there are always far more of those. treating a top all-time rank as proof of skill without accounting for how many people were rolling the same dice is the most common error in copy trading.
should i sort by followers instead?
for a different purpose, yes. profit rank tells you what a trader did. follower count is closer to a leading indicator of what their next entry does to a price, because on fomo the notification wave is a large part of the move. the two columns answer different questions and both are worth reading.
stop reading. start copying.
pick a trader from the fomo leaderboard, set your size, and the entries and the exits land in your own wallet while you sleep.
open copypump on telegram →